VladGor
Support Team
Joined: 26 Jan 2004
Posts: 528
|
 Posted: Tue Mar 09, 2004 4:34 pm Post subject: New version Genetic Optimizer 1.2 for TradeStation |
 |
|
TS.GO 1.2
Name: Trade Smart Research Genetic Optimizer for TradeStation
Version: 1.2 Full Release
Release date: Feb 28, 2004
Minimum system requirements:
Processor: 500 MHz Pentium III or better
RAM: 128 MB (512 MB RAM is recommended)
Hard Drive: 65 MB free space
Monitor: 800x600 monitor resolution or higher
Operating System: Windows NT/2000/XP
Software: Omega Research ProSuite 2000i or TradeStation 6.x/7.x
Technical characteristics:
- Max gene number = 1000.
- Max chromosome number = 1000.
- Max user variable number = 1000.
- Max population size = 1000.
New system features:
1. Now it’s possible to get an access to any unit of population. In previous version it was possible to get access to only the best unit.
| Quote: |
Prescription:
This feature allows to get TS results not for the best unit only but also for any unit within population. This might be useful for visual results analysis.
AlsoBy listing one system in several same windows and choosing representation of different units, we can estimate portfolio from these systems using TS portfolio.For this purpose we have to use GO_Get2 function instead of GO_Get. Example: Ind = Iff(GO_Next(Gen) = 1, NumExem, 0); Len1 = GO_Get2(1, Ind); .......WhereNumExem – number of unit, we want to see. During the test we get Ind=0 and then new candidate for test is chosen. On the last loop step Ind=NumExem, and we have results for required unit. |
2. User parameters for units are added.
| Quote: |
Prescription:
With any test unit we can relate additional parameters for an additional step description.Example:NetProfit, MaxDD, PF .... This allows us to save and view results related with specific unit. Also it’s possible to use description parameters for translating common data among signals within one strategy.
|
3.Access via parameters names is added.
| Quote: |
Prescription:
This feature allows building system from undependable units as it conducted in TS when strategy is built from signals. Each unit might has its own optimization parameters, which are defined within current unit. The parameters of different unit are undependable. |
4. The possibility of last loop step defining is added for signal generation and strategy report.
| Quote: |
Prescription:
Last loop step is defined in GO unit for displaying results for the best population unit. This flag might be used in EasyLanguage script, then strategy might behaves differently during an optimization process and displaying final result. This might be useful for Walk Forward Optimization performing.
|
|
|
